:You can check the stack frame for the RIP and if it is alreday in the
:kernel, you can skip the swapgs. Given that you have to do more work for
:traps from userland anyway, that's a relatively cheap change.
:
:Joerg
You can't, because not all traps to the system disable interrupts.
This means that another interrupt can occur after the stack frame has
been pushed, but before swapgs has been run the first time.
It really is a lot easier to just load a pointer from a fixed address.
-Matt
Matthew Dillon
<dillon@backplane.com>
| Alexandre Oliva | Re: Dual-Licensing Linux Kernel with GPL V2 and GPL V3 |
| Bart Van Assche | Re: [Scst-devel] Integration of SCST in the mainstream Linux kernel |
| Thomas Meyer | Re: [PATCH] clockevents: Fix suspend/resume to disk hangs |
| S.Çağlar | Rescheduling interrupts |
git: | |
| Chris Ortman | [FEATURE REQUEST] git-svn format-patch |
| Sverre Rabbelier | Git vs Monotone |
| Linus Torvalds | People unaware of the importance of "git gc"? |
| Johannes Schindelin | Re: VCS comparison table |
| Alexey Dobriyan | [PATCH 01/53] xfrm: initialise xfrm_policy_gc_work statically |
| KOSAKI Motohiro | [bug?] tg3: Failed to load firmware "tigon/tg3_tso.bin" |
| Jarek Poplawski | Re: Data corruption issue with splice() on 2.6.27.10 |
| David Miller | [GIT]: Networking |
| Nick Holland | Re: keyboard lockup, KVM, dual-boot |
| Richard Stallman | Real men don't attack straw men |
| Anders Langworthy | Re: OpenBSD/i386 won't boot on Transmeta Efficeon CPU |
| Matthew Dempsky | hoststated/relayd and Linux's tcp_tw_recycle option |
